Date: Thu, 4 Apr 2013 01:22:28 +0300
From: Sakari Ailus <sakari.ailus@iki.fi>
To: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Cc: Pali =?iso-8859-1?Q?Roh=E1r?= <pali.rohar@gmail.com>,
        Aaro Koskinen <aaro.koskinen@iki.fi>, Tony Lindgren <tony@atomide.com>,
        linux-omap@vger.kernel.org, linux-kernel@vger.kernel.org
Subject: Re: [PATCH v2] ADP1653 board code for Nokia RX-51
Message-ID: <20130403222228.GE10541@valkosipuli.retiisi.org.uk>
References: <201302171616.50152@pali>
 <201303241519.49243@pali>
 <514F7419.4010104@iki.fi>
 <2342477.1D1XZpYKqh@avalon>
MIME-Version: 1.0
Content-Type: text/plain; charset=iso-8859-1
Content-Disposition: inline
Content-Transfer-Encoding: 8bit
In-Reply-To: <2342477.1D1XZpYKqh@avalon>
User-Agent: Mutt/1.5.20 (2009-06-14)
Sender: linux-kernel-owner@vger.kernel.org
Content-Length: 4343
Lines: 102

Hi Laurent,

On Tue, Mar 26, 2013 at 12:07:01AM +0100, Laurent Pinchart wrote:
> On Sunday 24 March 2013 23:46:01 Sakari Ailus wrote:
> > Pali Roh?r wrote:
> > > On Thursday 07 March 2013 23:18:27 Sakari Ailus wrote:
> > >> On Wed, Mar 06, 2013 at 10:44:41PM +0100, Sebastian Reichel wrote:
> > >>> On Wed, Mar 06, 2013 at 09:20:16PM +0100, Pali Roh?r wrote:
> > >>>> On Wednesday 06 March 2013 21:12:06 Pali Roh?r wrote:
> > >>>>> On Sunday 17 February 2013 20:03:03 Aaro Koskinen wrote:
> > >>>>>> On Sun, Feb 17, 2013 at 04:16:49PM +0100, Pali Roh?r wrote:
> > >>>>>>> +/*
> > >>>>>>> + * arch/arm/mach-omap2/board-rx51-camera.c
> > >>>>>>> + *
> > >>>>>>> + * Copyright (C) 2008 Nokia Corporation
> > >>>>>>> + *
> > >>>>>>> + * Contact: Sakari Ailus <sakari.ailus@nokia.com>
> > >>>>>>> + *          Tuukka Toivonen <tuukka.o.toivonen@nokia.com>
> > >>>>>> 
> > >>>>>> You should put these people to CC... Just to see if the addresses are
> > >>>>>> still valid (which I doubt).
> > >>>>> 
> > >>>>> Ok, trying :-)
> > >>>> 
> > >>>> I got "Delivery Status Notification (Failure)" for both addresses.
> > >> 
> > >> This is expected.
> > >> 
> > >>> Sakari Ailus hosts some code on github [0], which has the following
> > >>> email address:
> > >>> sakari.ailus+gitorious@retiisi.org.uk
> > >>> 
> > >>> I added it to this mail's CC.
> > >>> 
> > >>> [0] https://gitorious.org/~sailus
> > >> 
> > >> Nice to hear people are interested in this. ;-)
> > >> 
> > >> The primary reason I haven't tried submitting this to mainline is that
> > >> ARM board code has a bad reputation these days. The N900 does not have
> > >> yet support for device tree (AFAIK), which also would require a few bits
> > >> and pieces on the flash driver to work.
> > >> 
> > >> Also the sensor and lens drivers would need at least some work before
> > >> being ready for submission to mainline for camera to be usable.
> > >> Unfortunately I haven't had recently time to work on this. N9(50) support
> > >> has higher priority for myself. That, too, is pending the DT support for
> > >> the device.
> > >> 
> > >> There's indeed more up-to-date code in my repository. Even if it's not
> > >> too close to mainline anymore it should be a better starting point than
> > >> the old kernel from MeeGo.
> > >> 
> > >> <URL:https://gitorious.org/omap3camera/pages/Home>
> > > 
> > > Hi,
> > > 
> > > this board code is same in your git repository and on meego obs.
> > > 
> > > Patch only adding support for adp1653 driver which is already in upstream
> > > kernel.
> > > 
> > > Are there any other problems with this patch to go for upstream?
> > 
> > A few more things comes to mind. We depend a little bit on actual board
> > code; it's not only static data. That's the configuration of the external
> > clock from the ISP to the sensor. That should move to the common clock
> > framework so that the ISP registers the clock and the sensor driver can then
> > use it. AFAIR Laurent has done some work on that.
> 
> Yes. I hope to get the patches in v3.10.

Cool! :)

> > The peculiar detail of the rx51 is that there's a switch on the camera CCP2
> > bus that selects either sensor (primary or secondary). Both sensors are
> > connected to the same receiver. That isn't properly modelled currently at
> > all, so that's why we have rx51_camera_set_xshutdown(). I guess it'd still
> > work if you only power (i.e. open) either of the devices at a time, though.
> 
> Have you thought about how we could model that ?

Well, the two dependent gpios could be modelled as two independent ones (
for sensor drivers), but setting the state of those gpios could fail,
gpio_set_value() still returns void. This isn't pretty perhaps but as a
result the initialisation of the secondary sensor to be powered up at the
same time will fail since it's in reset: the xshutdown of both sensors is
controlled by the same gpio as is the mux (AFAIR).

So one N900 camera specific gpio driver would be needed. It'd be a very
simple driver. What do you think?

-- 
Cheers,

Sakari Ailus
e-mail: sakari.ailus@iki.fi	XMPP: sailus@retiisi.org.uk
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/